Application of artificial neural network method to exergy and energy analyses of fluidized bed dryer for potato cubes

Drying the samples was performed in the inlet temperatures of 45, 50, and 55 °C, air velocity of 3.2, 6.8, and 9.1 m s⁻¹, and bed depth of 1.5, 2.2, and 3 cm. The effects of these parameters were evaluated on energy utilization, energy efficiency and utilization ratio and exergy loss and efficiency. Furthermore, artificial neural network was employed in order to predict the energy and exergy parameters, and simulation of thermodynamic drying process was carried out, using the ANN created. A network was constructed from learning algorithms and transfer functions that could predict, with good accuracy, the exergy and energy parameters related to the drying process. The results revealed that energy utilization, efficiency, and utilization ratio increased by increasing the air velocity and depth of the bed; however, energy utilization and efficiency were augmented by increasing the temperature; additionally, energy utilization ratio decreased along with the rise in temperature. Also was found that exergy loss and efficiency improved by increasing the air velocity, temperature, and depth of the bed. Finally, the results of the statistical analyses indicated that neural networks can be utilized in intelligent drying process which has a large share of energy utilization in the food industry. - Highlights: • Energy utilization increased by increasing temperature, air velocity and depth of the bed. • Exergy loss increased with increasing the air velocity, temperature and depth of the bed. • Prediction by a trained neural network is faster than usual mathematical models. • ANN it is a suitable method to predict the energy and exergy in various driers.